Vertical Transmission of Zika Virus in Aedes aegypti Mosquitoes

Introduction

Zika virus (ZIKV) is a mosquito-transmitted flavivirus in the family Flaviviridae. During the past decade, ZIKV has moved from the status of an obscure arthropod-borne virus (arbovirus) of little public health importance to the position of a Public Health Emergency of International Concern (PHEIC).1–3 Apart from reports of sexual and congenital transmission in humans, the primary transmission cycle of ZIKV is thought to involve primates (nonhuman as well as human) and certain species of Aedes mosquitoes.1,2 The available data indicate that there are two cycles of ZIKV; a sylvan cycle involving nonhuman primates and forest-dwelling mosquitoes, and an urban/suburban cycle involving humans and Aedes aegypti and, to a lesser extent, Aedes albopictus.1,2 In this regard, the sylvan and urban cycles of ZIKV are similar to those described for dengue virus (DENV), yellow fever virus (YFV), and chikungunya virus, the three other arboviral pathogens that are also transmitted by Ae. aegypti in their urban cycles.4–6

Since the declaration of the current Zika epidemic as a PHEIC, ZIKV research has focused mainly on the pathogenesis, genetics, molecular biology, and structure of the virus, with the goal of developing improved diagnostic methods, vaccines, therapeutics, and effective methods of disease prevention.3 In contrast, much less research effort has been directed at the mosquito component of the ZIKV life cycle, specifically factors affecting vector competence, transmission efficiency, and long-term maintenance of the virus.

One of the basic questions that has long puzzled arbovirologists is how arboviruses persist during adverse environmental conditions (cold periods in temperate regions and hot dry seasons in tropical zones) when adult vectors, such as mosquitoes, are absent or in very low numbers.7 Vertical (transovarial or transovum) transmission (VT) of a virus from female insects directly to their progeny is one mechanism for arbovirus maintenance in nature during adverse environmental conditions. VT can also maintain a virus in a specific locality, when most of the potential vertebrate hosts are immune, either as a result of vaccination or natural infection. There is both field and laboratory evidence that many arboviruses are vertically transmitted in their natural arthropod vectors, including some of the major flaviviruses such as DENV, YFV, Japanese encephalitis virus (JEV), St. Louis encephalitis virus (SLEV), West Nile virus (WNV), and tick-borne encephalitis virus.7 There is also evidence that many of the insect-specific nonpathogenic flaviviruses are also maintained by VT in their natural mosquito hosts.8,9

Given this background, we undertook laboratory studies to determine if ZIKV is vertically transmitted in Ae. aegypti and Ae. albopictus. This report describes results of our preliminary studies, demonstrating that ZIKV is vertically transmitted by infected Ae. aegypti females to some of their progeny, thus providing a potential mechanism for the virus to be maintained during adverse climatic conditions.

Materials and Methods

Results

Testing F₁ progeny for ZIKV infection.

A total of 69 pools consisting of 1,738 F₁ adult Ae. aegypti were tested. Six of the 69 pools were IFAT-positive (Figure 1), indicating that one or more of the F₁ progeny in the pool were infected with ZIKV. Based on these results, we estimated a minimum ZIKV filial infection rate of 1:290 for the experimentally infected Ae. aegypti. None of the 32 Ae. albopictus pools, comprising 803 adults, were positive in the IFAT.

View Full Size

Zika virus antigen in C6/36 cells inoculated with a pool of 25 infected F₁ Aedes aegypti adults, as detected by indirect immunofluorescent antibody technique.

Citation: The American Society of Tropical Medicine and Hygiene 95, 5; 10.4269/ajtmh.16-0448

• Download Figure
• Download figure as PowerPoint slide

Discussion

The observation that ZIKV was vertically transmitted by infected Ae. aegypti females to some of their F₁ offspring is not unexpected, in view of the results of other studies of VT with flavivirus pathogens in their mosquito vectors.14–27 Table 1 summarizes the results of selected published studies of VT of nine flavivirus pathogens by mosquitoes. In these studies, the parent female mosquitoes were infected by intrathoracic inoculation; although this is not a natural route of infection, the technique insures that all of the females are infected. In contrast, infection of mosquitoes by the oral route requires high titers of virus and the results are more variable. However, even in experiments in which 100% of the parent females were infected, the filial infection rates in their F₁ offspring varied widely, depending upon the mosquito species and geographic strain used, the virus type and strain used, the mosquito developmental stage tested (immature or adult), the virus assay method, the larval rearing temperature, the interval between initial infection and the first blood meal, and the ovarian cycle of F₁ offspring examined.14–27 For example, several studies16,17,20,22,25 have compared the filial infection rates of DENV, JEV, or SLEV in multiple geographic strains of a single mosquito species (i.e., Ae. aegypti, Ae. albopictus, or Culex quinquefasciatus) and have reported marked differences in the infection rates of their F₁ progeny. Similarly, other studies16,22–24 have compared the filial infection rates among different mosquito species after infection with a single flavivirus type. Alternatively, other studies16,17,22,26 have compared filial infection rates in a single mosquito colony experimentally infected with multiple strains of a single flavivirus type. In each case, considerable variation was observed in filial infection rates among F₁ progeny of the infected female parents. Two studies also reported that lowering the larval rearing temperature to 18°C instead of 27°C significantly increased the filial infection rates of mosquitoes infected with SLEV.16,23–25 Two other studies26,27 examined the filial infection rates among progeny of individual infected female mosquitoes, and these also varied widely. In summary, experimental studies of VT of flaviviruses in mosquitos have given highly variable results, indicating that there are multiple factors (variables) that can affect the frequency of VT and the resulting filial infection rates in female mosquitoes infected with flaviviruses pathogens. If such variability occurs in laboratory studies that are performed under relatively controlled conditions, then it must be even greater in nature. Consequently, one should interpret the results of such laboratory studies with caution, as they simply demonstrate that VT of these flavivirus pathogens can occur. But the critical question is “does VT occur in nature”? Answering this question is a much more difficult problem, as one would have to collect immature forms (i.e., larvae and pupae) or adult males of the vector species in an area where the virus of interest is endemic or epidemic and to demonstrate that they are infected.

Another potential variable is the presence of ISVs. The earlier experimental studies mentioned above and in Table 1 were performed at a time when most of the mosquito-specific flaviviruses were still unrecognized,8,10 consequently the infection status of the mosquito colonies used, with regard to mosquito-specific viruses, was unknown. Available evidence now indicates that mosquito-specific flaviviruses are common in nature and in laboratory mosquito colonies.8,10 These viruses are also maintained by VT in their insect hosts and at much higher filial infection rates than the flavivirus vertebrate pathogens.8,10 The effect of these ISVs on the vector competence of their mosquito hosts is uncertain and may represent another important variable affecting VT of flavivirus arboviral pathogens, such as ZIKV, DENV, YFV, JEV, SLEV, and WNV.8,10 One potential effect of prior infection of a mosquito with an insect-specific flavivirus is superinfection exclusion (or homologous interference), a process by which cells infected with one virus do not support replication of the same or a similar virus.9 This phenomenon has been observed during infection of both vertebrate and invertebrate hosts with a broad range of viruses, including some of the mosquito-specific flaviviruses.9 Consequently, the absence or presence of mosquito-specific flaviviruses in a laboratory mosquito colony potentially could also alter the VT rate of a flavivirus pathogen in that colony.

In view of the relatively low filial infection rates observed with ZIKV and other flavivirus pathogens (Table 1), one might conclude that VT is of little epidemiologic importance. With the low filial infection rates observed in those experimental studies, mathematical models28,29 predict that flavivirus pathogens such as ZIKV, DENV, YFV, or WNV would survive for only a few generations without horizontal transmission and amplification in a vertebrate. However, VT rates in this range may be sufficient to allow a flavivirus to persist during hot dry periods or cold weather, when adult vectors are absent or in low numbers.29 Many Aedes species, including Ae. aegypti and Ae. albopictus, produce resistant eggs that can survive for months (or longer) in a dried dormant state.30,31 Although adults may not survive a winter or dry season, when favorable environmental conditions return, the eggs hatch and larvae emerge to complete the insect's life cycle and to establish a new generation. As the larvae pass through various developmental stages and grow, the virus also replicates, so that by the time an infected adult female emerges, she is infectious and able to transmit the virus.

Venereal transmission is another mechanism by which a virus can be amplified in a mosquito population. Although male mosquitoes do not take blood, they can acquire virus by VT from an infected female parent. In experimental studies, infected male mosquitoes can transmit virus horizontally to noninfected adult females during mating as well as to her developing oocytes, resulting in infected F₁ progeny.7,32

Another mechanism for virus to be maintained in an insect population is the stabilized infection model, as described with Sigma virus (Rhabdoviridae) in Drosophila melanogaster.33 In this condition, the virus is maintained by VT at a high level in nature by relatively few infected females. Stabilized infections have been demonstrated in several Aedes mosquito species with California encephalitis serogroup viruses (Orthobunyavirus: Bunyaviridae).34,35

Because of these alternative methods of virus transmission, the importance of VT should not be discounted, based solely on low filial infection rates observed in experimental laboratory studies.

Knowledge of the transmission mechanisms of arboviruses in their vertebrate and invertebrate hosts continue to evolve. ZIKV was once thought to be transmitted to humans solely by the bite of infected mosquitoes. The recent pandemic in the South Pacific and the Americas has shown that other modes of human infection can occur (venereal, congenital, postpartum, and blood transfusion).1,2 The classical concept that arboviruses are maintained in nature by continual transmission between susceptible vertebrate hosts by hematophagous arthropods is also outdated and incomplete, as alternative mechanisms for virus maintenance within the vector population have been described. ZIKV is probably no exception. As a survival strategy, a successful parasite (in this case an arbovirus) might be expected to have alternative maintenance mechanisms to insure its survival during periods when susceptible vertebrate or arthropod hosts are not available. The alternative maintenance mechanisms (survival strategies) for ZIKV in its mosquito hosts is a neglected area of research, but this information is essential if we want to fully understand the ecology of ZIKV and to eventually control it.